Recent years have seen substantial advances in understanding of the neurobiology of goal-directed decision making. There is evidence that such decisions produce specific neural activation patterns hypothesized to reflect encoding of subjective or goal values (GVs), defined as the predicted reward value of available options, calculated in a 'common currency' at the time of decision14,15. Many of these gains in knowledge are emerging from the discipline of neuroeconomics. Despite the clear relevance of these findings for drug dependence, arguably pathology of decision-making16, 22, little is known about the neurobehavioral mechanisms of the most clinically-relevant decisions made by drug users: decisions to obtain and use drugs. This application proposes two parallel within-subjects studies examining the neural encoding of GVs during choices between money and marijuana (MJ~ Study 1), cocaine (COC~ Study 2) and palatable food (both studies) in regular MJ and COC users, and the impact of cue exposure on these processes. Employing a combination of human behavioral pharmacology, advanced fMRI methods, and neuroeconomic approaches, we aim to contribute an understanding of the neurobehavioral mechanisms underlying decisions to use drugs versus non-drug rewards in regular drug users, knowledge that could ultimately play a central role in developing interventions to prevent these decisions. Regular non-treatment-seeking MJ (e4x/week~ N=20) and COC (e2x/week~ N=20) smokers will complete a 4-day inpatient protocol including 4 randomized conditions: 1) Drug versus money choices after neutral cues~ 2) Drug versus money choices after drug cues~ 3) Food versus money choices after neutral cues~ and 4) Food versus money choices after food cues. A range of choices between drug/food and money will be made in an fMRI task, after which one choice will be randomly selected for implementation, with MJ, COC or food administered accordingly. We aim to 1) characterize neural mediators of relationships between the amount of drug or food offered and the decisions made~ 2) Investigate differences in behavior and neural GV signals during choices about drug compared to choices about food~ and 3) Compare effects of drug and food cues on behavior and GV signals during choices about drugs relative to food. We anticipate that signaling in the ventromedial Prefrontal Cortex (vmPFC) and striatum will mediate relationships between the amount of all three rein forcers offered and decisions made. We further hypothesize that drug-related decisions, and decisions after drug but not food cues, will involve quantitatively greater GV signaling in vmPFC and striatum, thought to be the final seat of valuation in the primate brain14,15. Execution of these studies will be combined with a comprehensive training program that includes didactic training in Matlab programming, advanced fMRI analysis, and neuroeconomics, guided readings in human behavioral pharmacology and fMRI methods, and supervision and oversight by a team of internationally- recognized experts in human behavioral pharmacology (Drs. Haney and Foltin), advanced fMRI analysis (Dr. Lindquist), and the neurobiology of decision-making (Dr. Glimcher). The well developed training plan proposed will therefore provide the Principal Investigator, Dr. Gill Bedi, with the advanced-level skills in these areas to initiate an independent research program, using human behavioral pharmacology and advanced fMRI skills to investigate clinically-important questions about drug addiction. Dr. Bedi has graduate training in clinical psychology and postdoctoral training in human behavioral pharmacology involving illicit drug challenge studies in non-clinical, but not drug-using, populations. She has a strong track record of publication for an early career investigator, and has demonstrated capacity to obtain NIH funding for small grants, having been awarded an R03 and a R21 in 2011. While this emerging research track bodes extremely well for her capacity to develop into a creative and productive independent researcher, the protected time, training, and mentored research experience that would be provided under this K23 award are critical at this juncture of her career to allow her to complete the transition to full independence as a patient-oriented researcher. Dr. Bedi's immediate goal is to develop advanced skills in human behavioral pharmacology and fMRI analysis that would allow her to obtain independent R01 funding. Her longer-term goal is to develop a programmatic line of research employing these methods within the Substance Use Research Center in New York State Psychiatric Institute and Columbia Psychiatry. The Institutional environment available to support Dr. Bedi's development as a researcher is among the best in the world, combining ready access to internationally-renowned senior faculty, an established history of successful development of young researchers, and world-class facilities including medical facilities to support studies involving marijuana and cocaine smoked self-administration and a 3T research-dedicated GE MR scanner housed in the NYSPI MRI Center. Thus, this application proposes an innovative project with the potential for substantial impact. It combines expert mentorship and comprehensive training to facilitate the transition of an early career researcher with a promising emerging track of publication and small NIH grant funding to full independence as a patient-oriented researcher.